An extended generalized filter algorithm for urban expressway traffic time estimation based on heterogeneous data

Travel time estimation and its variation for urban expressways are vital to both the information provision to road users, and the system evaluation and management for traffic administrators. Fruitful research efforts have been made to develop methodologies of reconstructing spatiotemporal traffic states mainly for freeways based on one or multiple data sources. However, few studies specifically focused on urban expressways. There are more intensive merging and diverging traffic due to short distances between ramps, for example, 300–500 m. Based on the empirical analysis of traffic data collected on a typical segment of a congested urban expressway, this study proposes an extended generalized filter algorithm for the urban expressway traffic state estimation based on heterogeneous data. More specifically, the multiple sources of data include both fixed sensor data (e.g., inductive loops or radar data) and global positioning system (GPS) probe vehicle data. This study compares the proposed algorithm and the traditional algorithm for freeways using data collected on the segment of expressway in Beijing, China. Results demonstrate the advantage of the proposed method, as well as its feasibility and effectiveness.